This replication repo contains data and code for the paper 

"Political partisanship influences behavioral responses to governors' recommendations for COVID-19 prevention in the United States" PNAS, 2020. 

by Guy Grossman, Soojong Kim, Jonah M. Rexer, and Harsha Thirumurthy. 

Inside this repo there are two sub-directories:

1) code:
- figures_Stata.do: Contains the code to replicate Figures 2, 3, and SI-6 - SI-12 using Stata.
- figures_R.R: Contains the code to replicate Figures 1, SI-1, SI-2, SI-3, SI-4, and SI-5 using R. 
- tables.do: Contains the code to replicate all Tables using Stata. 

To run the do-files on your machine, set the appropriate paths for the globals at the top of the file. 

2) data:
- SG_county_data.dta: County-level labeled daily data on mobility, governors' messaging, partisanship, and other covariates. Input to figures_Stata.do and tables.do
- gtrends_dma_data.dta: DMA-level labeled daily data on google trends, governors' messaging, partisanship, and other covariates. Input to figures_Stata.do and tables.do
- precint_data.dta: Precinct-level daily data on mobility, messaging, and electoral returns. Input to figures_Stata.do and tables.do
- r-figures-data.RData: data on the timing of governors' communications for the R script figures_R.R


